Combating Ambiguity for Hash-Code Learning in Medical Instance Retrieval

• Main Authors: Fang, J. (Author), Fu, H. (Author), Liu, J. (Author), Yan, X. (Author), Yan, Y. (Author), Zeng, D. (Author)
• Format: Article
• Language: English
• Published: Institute of Electrical and Electronics Engineers Inc. 2021
• Subjects: accuracy; algorithm; Algorithms; Alzheimer disease; Article; Classification (of information); classification algorithm; Codes (symbols); content-based image retrieval; Convolution; convolutional features; cryptococcosis; Databases, Factual; deep hashing methods; Diagnosis; diagnostic imaging; Discriminative features; electroencephalography; entropy; factual database; Feature aggregation; glaucoma; Hash functions; hematoma; human; Humans; Image coding; image reconstruction; image retrieval; image segmentation; Instance retrieval; Large image database; learning; learning algorithm; machine learning; Medical imaging; Medical instance retrieval; methodology; Military photography; optic disk; Query processing; radiologist; Radiologists; receptive field; Research Design; Retrieval performance; Semantic information; semantics; Semantics; Spatial differences; State-of-the-art methods; support vector machine; training; visual field
• Online Access: View Fulltext in Publisher

 When encountering a dubious diagnostic case, medical instance retrieval can help radiologists make evidence-based diagnoses by finding images containing instances similar to a query case from a large image database. The similarity between the query case and retrieved similar cases is determined by visual features extracted from pathologically abnormal regions. However, the manifestation of these regions often lacks specificity, i.e., different diseases can have the same manifestation, and different manifestations may occur at different stages of the same disease. To combat the manifestation ambiguity in medical instance retrieval, we propose a novel deep framework called Y-Net, encoding images into compact hash-codes generated from convolutional features by feature aggregation. Y-Net can learn highly discriminative convolutional features by unifying the pixel-wise segmentation loss and classification loss. The segmentation loss allows exploring subtle spatial differences for good spatial-discriminability while the classification loss utilizes class-aware semantic information for good semantic-separability. As a result, Y-Net can enhance the visual features in pathologically abnormal regions and suppress the disturbing of the background during model training, which could effectively embed discriminative features into the hash-codes in the retrieval stage. Extensive experiments on two medical image datasets demonstrate that Y-Net can alleviate the ambiguity of pathologically abnormal regions and its retrieval performance outperforms the state-of-the-art method by an average of 9.27% on the returned list of 10. © 2013 IEEE.